On the use of planar Doppler velocimetry

Velocity measurements using the Planar Doppler Velocimetry (PDV) technique were made in a small jet facility and in a large wind tunnel to study the feasibility of the technique including sources of error and means to minimize them. PDV is a non-intrusive technique which allows the measurement of one or more velocity components everywhere in a plane within a flow field. Its potential to yield instantaneous, simultaneous velocities throughout a plane and at high spatial resolution in a single measurement makes it a promising tool for turbulence studies. PDV measurements are made by detecting the Doppler shift produced when laser light is scattered from moving particles in a flow field. The current PDV system utilizes a pulsed, injection-seeded, frequency-doubled Nd: YAG laser capable of producing narrow linewidth light (~140 MHz) that can be tuned to frequencies associated with the absorption lines of iodine. An iodine filter may therefore be used to discriminate the frequency of the laser light. A system utilizing two scientific-grade CCD cameras was used to record images produced by the scattered laser light. An unusual feature of the experimental arrangement was a frequency monitoring system which tracked the laser set point frequency on an image to image basis, providing resolution to better than 4 MHz. In a Mach 1.36 free jet, mean flow velocities in the core (averaged from three streamwise stations) were measured by PDV to within 2.4% of the value obtained by LDV (to within 6.4 m/s). For the high speed core of the jet, instantaneous velocity fluctuations greater than 3.4% (about 9 m/s) were measurable by PDV, but lesser fluctuations were masked by PDV system noise. In the wind tunnel (3.05 × 2.13 m test section), uniform flow runs at 68 and 96 m/s indicated bias and random errors of less than 2 and 4 m/sec, respectively. Mean measurements of the flow over a delta wing at a 23° angle-of-attack were similar to the results of a CFD calculation by Rizzetta (1996). The dominant source of random error in the velocity measurements arose from the phenomenon of laser speckle and the dominant source of bias error came from the characterization of the iodine filter profiles.